Setting up on local (Microsoft-Word web)

• Open a Microsoft word document on web
• Navigate to insert tab
• Click add-ins button, upload my add-in
• Drag add-in/manifest.xml into the upload bar
• Click upload

If you can't find the Add-ins tab, look instead on the File menu for "Get Add-ins" or something like that, then click Manage Add-ins.

Running the add-in

• cd frontend
• npm install
• npm run dev-server

Run npm run lint --fix to auto-fix (almost) all lint errors.

Running the backend server

After cloning this repo:

1. uv sync in the top-level folder; activate the .venv environment that this creates
2. cd backend (for all the following steps)
3. Run uv run get_env.py to create the .env file. (It'll prompt you for an OpenAI API key.)
4. Run the server: uv run uvicorn server:app --host localhost --port 8000 --reload

Run ./test_generation script in the backend folder to make a test request. However, it runs against the prod server; change the URL to run against a local server (e.g., http://localhost:8000/api/generation).

Note: the custom LLM backend has moved to https://github.com/AIToolsLab/writing-prototypes.

Contributing

We follow specific GitHub conventions to keep our project organized and maintain code quality.

Quick Start:

1. Create an issue for your task
2. Create a branch: <type>/<description> (e.g., feat/add-user-login)
3. Make commits: <type>: <description> (e.g., feat: add login form)
4. Open a PR and link to your issue
5. Get at least one approval before merging

For detailed conventions on branch naming, commit messages, PR process, and issue management, see CONTRIBUTING.md.

Running visual regression tests

The visual regression tests capture screenshots and ensure UI consistency of the application. See VISUAL_REGRESSION.md for detailed instructions on running tests and updating baseline images.